http://wecott.csail.mit.edu/

WeCott is a (prototype) website built by myself and three friends during the iCorruption hackathon a few weeks ago. We envisioned a platform of communities where people could participate in boycotts together – offering advice on alternatives, uploading photos of their boycott, and otherwise supporting each other. Not only would this make the process more fun and supportive, it also allows people and companies to see the effects of the boycott all in one place. You can see some more of our thoughts and designs in our presentation pitch.

These last two weeks, myself, Giovana, Alicia, and Wahyu have been brainstorming how to use this platform to talk about issues that we cared about. The first thing that we realized was that while boycotts could be very powerful as a general tool, it’s not often clear exactly what to boycott given a particular issue. Sometimes problems are so systemic (such as Ferguson) or involve things that we are unable to give up (such as our government). These types of problems may involve some more creative thinking about what could be done as an individual. Other problems fit well into the boycott mold. However, we acknowledged that often boycotts hurt people lower down the power ladder such as workers instead of the business leaders we want to target. At the end of the day, that is a tradeoff that we must make as individuals when voting with our money. Sometimes, whether we know it or not, we can be perpetuating or contributing to entities and problems that we do not support, just when we go to the grocery store!

Giovana and Wahyu both contributed a petition to WeCott – here and here. They can speak more about the specific issues they investigated and the thought processes they went through to create the petition.

I can talk a bit about the design of WeCott and also what else I envisioned but didn’t get to prototype. As you can see, there is a dollar amount that is associated with each petition that is supposed to collect all the money that people have pledged that they will divert. A bit about this – I felt that having a dollar amount would be a really powerful signal to companies. And while we can’t guarantee that people will stick to that, we have thought about ways to help people that pledge, such as sending notification emails, and to encourage people to be accountable, such as uploading photos or maybe checking in on a map to then gain points or badges. I can see other features such as news updates, goal setting, discussion boards, and local or map integrations that could help keep momentum going and provide local support. A lot of the initial design inspiration for WeCott came from both Change.org and Kickstarter. One member of the original team will be presenting WeCott again at the bigger iCorruption conference some time next week.

I did most of the work in python, using the awesome libraries of tweepy (Twitter API wrapper), matplotlib (plotting), pymongo (interface to mongo database), and nltk (natural language toolkit). I used a mongo database to store the data but it wasn’t super necessary (plain text files can easily suffice). I forgot to take into account how long it would take for the scripts to crunch through all the data, so when I got started last night, I quickly realized I’d better let the scripts run overnight and write up a post this morning.

My dataset consisted of 663131 tweets related to abortion collected from the year 2013. To find tweets related to abortion, I looked for key terms such as “abortion”, “abort” + “baby”, “abort” + “birth”, “prolife”, “prochoice”, and some others, including common hashtags.

Here is some basic info on the tweets I collected:

Total Volume of Tweets over time (x=month of 2013, y=number of tweets):

You can see that the volume varies quite a bit. Looking at the top words used each month, removing stopwords (very common English words), we see the following (I show the word as well as the number of times that word appeared in tweets in that month):

We can clearly see that some volume seems to be driven by news events, such as Senator Wendy Davis’s filibuster in June to block a restrictive abortion bill in Texas. Other drivers perhaps include Twitter campaigns (#praytoendabortion). This also is a good point at which to audit one’s data and zoom into weird findings to check if the data is properly cleaned. I didn’t have time to do that here, but if I did, I would look at the tweets behind some of the weirder top 5’s that I didn’t understand and either learn something new about abortion or find ways to remove the invalid tweets from the dataset.

The last thing I did was to analyze the language in the tweets for moral values. This is part of a larger research project I am working on related to modeling ideology and linking that to policy change. You can see a complete version of this work here when I looked at same-sex marriage. Another important step which I am skipping is validation, or trying to correlate numbers crunched from the data to traditionally collected data, such as census or poll numbers.

To analyze moral values, I am using a supplemental LIWC dictionary built by political psychologists and linguists that attempts to match key words with underlying moral values. The 5 moral values we use are taken from research on moral foundations by political scientist Jonathan Haidt and some other people. They’re an attempt to understand the underlying values that people find important. Do you care more about fairness or more about loyalty and authority? Not surprisingly, these moral foundations somewhat correlate with either liberal or conservative ideologies.

So, given the keywords ascribed to each moral foundation, I counted the relative occurrence of each moral foundation within every tweet and then averaged that relative occurrence across all tweets within a month. The following is the outcome:

Moral Values Over Time (x=month in 2013, y=average relative occurrence in tweets)

Though we see more authority and harm language than the other 3, this doesn’t necessarily mean that people think more about some values relative to other values because our method can’t be comprehensive. But we can look at a single value over time. For instance, it’s pretty notable how purity language jumps up in October. I didn’t have time to dig into why but that would be my next step.

Future work that I intend to do would be to look at these traits broken up by state. You can do that by analyzing the location field that people specify in their profile and trying to match that to a state. I would want to look at several of the other LIWC categories and also come up with some more features of my own. Finally, it would be interesting to look at features over time – leading up to and after key events, for instance.

The issue that we chose to focus on was suggested by Laurie, who had a strong interest in the topic. We focused on the fallacies that the incumbent Conservative party in Britain have been stating in order to paint the last 5 years of their governance in a positive light as elections loom. We wanted to combat these statements by showing how policies enacted in the last several years have actually made many people’s lives worse off, except for those in the top percentages.

We wished to make the presentation interactive to allow users to take a part in shaping the information they receive, which may make them more receptive to the information that is presented. We also strove to frame the quiz at the outset in a neutral way. We added interactivity by building a quiz that asks the user about their circumstances. The answer to the question “Have the last five years been a recovery or crisis?” depends on the circumstances that a user puts in. We didn’t have time to completely finish the quiz as of this post, so many pertinent questions are currently omitted and the answers are only partially written. Ideally, we would also have a meta-analysis at the bottom that allows you to quickly see how your results compare to other people’s results.

The quiz is available here: http://people.csail.mit.edu/axz/quizlet/quiz.html

It is written using HTML, CSS, and Javascript. The code is available here: https://github.com/amyxzhang/quizlet

I knew a tiny bit about Austin Hess beforehand from listening up in class – mainly that he was editor-in-chief of the MIT Tech, and thus I presumed that he was an MIT senior, probably around 21.

When I sat down to Internet stalk Austin Hess, I first came across what I thought was his Twitter. The handle was @AustinHess, the photo was of a youngish male, and the location stated Boston, MA. Not bad huh? Everything checked out, so I scrolled down the tweets, excited that I was going to get a glimpse into his inner psyche.

Reading his tweets, this Austin Hess seemed something like a raging Tea Party affiliate with hashtags like #EmperorObama and um, #oldmanass. I began imagining Austin in my head, this 30-year-old, angry, married, but also an MIT student working at the Tech. I was kind of looking forward to interviewing someone who seemed like my direct opposite in every way.

A few more Google searches and I realized I was totally off base. The MIT Austin was someone else entirely. He unfortunately was not nearly as public with social media as Tea Party Austin but I was able to dredge up some old articles about his high school career. From what I could find of him in high school, he seemed like he was a really bright, accomplished person and really into science, specifically physics and space (the kind of kid that would get into MIT!)

Then in college, I dig some digging and found a short blog he wrote while interning at CERN when the Higgs Boson discovery was announced. I found all his articles that he wrote for the Tech, including some recent ones regarding the Walter Lewin case at MIT. …And that was it!

So it was on to the interview to get to know more about him. I went to the Tech offices on the 4th floor of the MIT student center at around 5PM on a Saturday. While the floor was basically empty, behind the locked doors of the Tech were over 20 people in various rooms talking and working. After finally locating Austin, we found an empty room and began talking.

Austin talked about being editor-in-chief of the Tech. He enjoyed controversial stories and hearing people’s back-and-forth on it. One of his bigger, controversial stories was on Walter Lewin, who he even had some back and forth with over email. “It was a really surreal experience. Some of the email conversations were very strange. And at the end he got very mad that he still wasn’t portrayed in a positive light.” Huge swaths of people got really angry at the Tech and wrote comments and emails on a story that he initially thought was quite straightforward.

One of the former editor-in-chiefs told Austin before his tenure, “This is going to be a crazy time but you’ll learn more than you will in any MIT class.” Though Austin doubted him at the time, he in the end found it a learning experience: learning how organizations work but also how complicated issues can get between what’s in the news and what actually happens. It also made him a little disappointed in the MIT community from time to time, because he was often on the receiving end of what he considered to be extreme voices.

Austin also talked about his time interning at CERN during the announcement of the discovery of the Higgs Boson, which was a unique and very exciting experience. He remembered all the practice announcements, arguments between scientists, and staying up all night in line waiting for the final big announcement.

While it was a definite high point, possibly one of the greatest high points for physics for many years to come, this also made him realize during this time that he wasn’t that interested in pursuing physics research further. He saw that many post-docs and graduate students working on the Higgs Boson, a truly tremendous project, were still leaving the field and leaving research altogether just several months later.

Finally, he told me about his plans for the future, including most likely (*maybe, maybe not) working at the New York Times as a software engineer, a job that would combine his interests in technical subjects and the news in a city where things are happening all the time! Looks like you have a really bright and exciting future ahead, Austin!

The new Ed Portal

Despite the freezing temperatures and light snow on the afternoon of Saturday, February 23^rd, the grand opening of the Harvard Ed Portal in the Boston neighborhood of Allston was a lively and cheerful affair. Well over 200 people attended the event from 1-4pm, along with several members of the Harvard and local city press, with organizers and students from Harvard University mingling with families from the Allston-Brighton community.

The new Science Room at the Ed Portal

Both Harvard President Drew Faust and Mayor of Boston Martin J. Walsh were in attendance and gave short speeches in recognition of the event. The afternoon also featured musical performances and dances by various student groups, and an enthralling lecture on the piece “The Rite of Spring” by Harvard Professor Tom Kelley from his upcoming HarvardX course. For the children, there were painting exercises, fun science experiments, and other activities led by Harvard student volunteers.

There were lots of children and many activities for them to take part in.

Professor Tom Kelley gives a lively demonstration of the dances and music in the Rite of Spring.

The Harvard Ed Portal is a brand new 12,000-square-foot space at the corner of Western Ave and North Harvard Street in Allston intended to serve as a community center and place of learning for the residents of Allston and Brighton. The space contains a theater space for performances as well as many smaller rooms for workshops and activities for children.

Entrance to the new Ed Portal in Allston.

This grand unveiling was the latest in a series of efforts from a task force led by Professor Rob Lue, faculty director of the Ed Portal, to extend the many resources available at Harvard University to members of the Allston community. In total, Harvard has allotted $8.3 million towards the building of the Ed Portal and other projects aimed at the local community. Many Harvard students have also taken on roles as organizers to carry out community projects and serve as mentors to children.

President Drew Faust addresses the packed audience.

At the new theater podium, speaking to a mostly standing audience, President Faust asked attendees to consider the question, “What is a Harvard? How do we think about what Harvard is?” She then went on to paint a picture of Harvard as “the unending pursuit of knowledge” and the Ed Portal as a place where the community around Harvard could take part in this pursuit as well.

A sign welcoming President Faust and Mayor Walsh.

Mayor Walsh echoed the question again in his speech and stated, “To me growing up, Harvard was someplace that very, very smart people went to, got a good education, and went on to do great things…went on to become presidents, and kings, and prime ministers. Today, what is a Harvard to a lot of people in this room is that Harvard is in reach of every person in this room.”

In his speech, Mayor Walsh acknowledged the sometimes-strained relationship between Harvard University and its neighboring communities as well as the city of Boston when he said, “I want to thank Harvard. I know that sometimes there’s been a bit of bickering back and forth. But this is one of the things that come out of bickering. It’s a great opportunity.” The frank statements elicited a lot of knowing chuckles and applause from the audience, many of whom were residents of the neighborhood.

The statements alluded to a long history between Harvard and Allston reaching back a quarter of a century when Harvard began buying up land in Allston across from the Charles River. While Harvard Business School and the new Innovation Lab are already in Allston, many new developments are planned as Harvard expands into the neighborhood, including a new home for the School of Engineering and Applied Sciences and potentially new residential spaces for students.

Many residents of Allston have been upset about this encroachment of Harvard into their community and concerned about what the continuous march of development means for them. Many have complained of how this has resulted in higher living costs for current residents, pricing many of them out of the neighborhood.

One sign of this was seen in an advertisement found outside the Ed Portal promoting the Continuum, a brand new retail and apartment complex across the street. It had the word “gentrification” scrawled across the front. It remains to be seen how Harvard’s plans will continue to change the face of Allston and its residents in the coming years. However, the Ed Portal, as Mayor Walsh stated, has been one good outcome of the tensions between the university and the city and will hopefully be a entertaining and inspiring place for the community for years to come.

————

My Takeaways from this experience:

Attending the event was a somewhat stressful experience, as I tried to juggle food and drinks in one hand and my phone in the other, which I was using to both record audio and take pictures. My audio quality was terrible as I couldn’t get close enough to the stage and was too close to the rooms of children. I think being an actual journalist may be difficult for me as I am by nature quite non-confrontational. Although, I imagine some credentials and a badge could make me a lot more willing to push past crowds and even try to interview folks. I did not attempt to interview anyone for this piece.

I also didn’t have time directly after the event to write up the article and so cheated and wrote it several days later (still taking up only 4 hours in total for event + writing). Trying to coordinate WordPress and images on my iPhone turned out to be extremely frustrating due to the opaqueness that is iPhoto Library app. It’s amazing to me, even as someone who does UX and usability, how much anger something like poor or purposefully obfuscated application design can elicit.

Users create an account on the website and install a Chrome extension. They can then choose to whitelist certain domains or selectively publish their visits to certain pages, which the extension tracks and pushes to their feed. You can also do other things like see who else has been on the page you are on, post comments or chat on any page on the web, or follow other people to see visits from their feed.

Here is a link to my feed: http://eyebrowse.csail.mit.edu/users/amyxzhang

And here is a screenshot:

My eyebrowse profile, showing my most recent shared browsing activity.

As you can see, my profile contains the webpages I’ve chosen to share in reverse chronological order, links to visit them, ability to filter by keywords and time, and tags added by myself. There’s a public API for anyone to play with the data themselves (http://eyebrowse.csail.mit.edu/api/v1/history-data?format=json&user__username=amyxzhang&offset=0&limit=10). There’s also a visualization page containing some dynamic visualizations with the ability to filter, save as static image, and embed as a widget on a webpage (http://eyebrowse.csail.mit.edu/users/amyxzhang/visualizations?query=&date=last%202%20weeks)

Unfortunately, wordpress.org does not allow code snippets so I can’t embed the dynamic widgets. You’ll have to visit the webpage above for those. However, here are some static images of my activity over the last two weeks:

A word cloud of page titles from webpages I visited in the last two weeks.

My browser visits broken down hour and by my top domains over the last two weeks.

My browser visits broken down by day of the week and by my top domains over the last two weeks.

One can clearly see that I spend quite a lot of time on coding websites, especially this last Monday, when I decided to spend the holiday upgrading this very application from Bootstrap 2 to 3.

Since I’ve been collecting my browsing data for a long time now (almost a whole year I believe), I can also go much further back to notice larger trends. Here are those last two graphs again except over the last year:

My browser visits over the last year broken down by time of day.

It seems that I’ve been getting better about stepping away from the computer before 2AM, at least in the past two weeks compared to the last year. And since activity on StackOverflow is a good indication that I am coding, it seems that I’ve conducted a lot of this very late at night. This graph also makes painfully clear how late I start the day on average. I’m also surprised by the presence of Mashable in the top 10, as it’s a media site I never really thought I visited often.

My browser visits over the last year broken down by day of week.

Looking at days in the week, it’s interesting to see that my coding work (or visits to StackOverflow) increases from its lowest point on Monday, reaches a peak on Thursday, and quickly tapers off once I hit Friday. My media consumption however remains fairly steady.

Now for some more high-level reflections of this whole experience. Before two weeks ago when I got this assignment, I had only whitelisted certain web domains that I was reasonably comfortable sharing with the world – things like Wikipedia, the New York Times, and various coding and research related websites. Beginning two weeks ago, in an effort to capture more of my media diet, I started whitelisting everything that remotely resembled news or media (excluding social media – I still wasn’t comfortable sharing that), so all my embarrassing visits to BuzzFeed and random gossip sites were also tracked and shared. The experience was really interesting to me not just to see what I had visited and notice trends, but also in a meta way to see how my tracking and sharing of my browsing history caused me to browse differently. Particularly this experience made me much more mindful of my media consumption and careful and picky about how I chose to spend my time online. For instance, in Eyebrowse, there’s a feature that makes it possible to at any moment turn off all tracking on even whitelisted domains, effectively turning off Eyebrowse as if it were in incognito mode. By having that option available, I became more thoughtful and aware of what I was doing when I had the option to choose to be privately or publicly browsing. And while tracking and visualizing the data played a role in that, it was the added step of then having that data be shared and choosing when and what data to share that made me very conscious of my bad (and good) habits.

While I hadn’t explicitly developed Eyebrowse for this purpose before, this experience has made me think of the potential benefits of a social app like Eyebrowse towards not just monitoring but keeping accountable one’s goals for their information diet. Indeed many current applications related to maintaining exercise habits and food diets incorporate social sharing to provide a measure of accountability and support. And after all, why shouldn’t we be as mindful of what we feed our minds as we do our bodies? If you have any thoughts around this, I’d love to hear it! As I continue to develop Eyebrowse, I will work on adding features to make this process easier, included better aggregate statistics and visualizations and easier ability to share these reports (like little media diaries!) on social media and personal websites. I’ll also build in more levels of obfuscated sharing, for instance the ability to share that I’m on Facebook but not specific pages.

If you’re interested in seeing where this goes, I encourage you to try out Eyebrowse! I’m still actively developing it and would love to get some more users as well as any feedback and bug reports. I haven’t released it to the public or anything yet – just publicized it around MIT CSAIL and my research group – so the only people using it right now consistently are me and my advisor, David Karger. By default, nothing is tracked when you install the extension. If you have more questions, feel free to contact me (axz@mit.edu) or read our FAQ: http://eyebrowse.csail.mit.edu/faq.